Mechanical, physical or chemical recycling: compare the same job

A comparison between recycling routes is valid only if three conditions hold. Where any of them fails, the comparison is measuring the difference between the routes' assumptions rather than between the routes.

The first is the same input specification. Mechanical, physical and chemical routes accept different inputs, and their suitability is partly defined by that. Comparing a route designed for clean, single-polymer, well-sorted input against one designed to handle mixed or contaminated material is not comparing like with like, because the routes are doing different jobs. A valid comparison either fixes the input and assesses which routes can handle it, or fixes the output and assesses which routes can reach it from the available input.

The second is the same functional output. The routes produce different things: mechanical recycling produces a recyclate that can be re-formed; physical processes produce a purified material; chemical processes produce monomers, feedstocks or fuels that may displace virgin material at a different point in the value chain. Comparing a route that produces a material equivalent to the original against one that produces a fuel is comparing two different outputs, and the comparison should say so rather than aggregate them under a single measure.

The third is a consistent system boundary. What is counted as input, as output, as energy and as residue must be defined the same way for each route. A route credited with displacing energy, for instance, is producing a different kind of output from one producing material, and treating them within one boundary without stating the difference produces a comparison that cannot be interpreted.

The routes differ materially in how established they are at commercial scale, and that difference belongs in the comparison rather than being treated as a footnote.

A proven route at the scale a project needs carries a different risk profile from one demonstrated at pilot or demonstration scale. The difference is not only technical feasibility: it appears in the guarantees a supplier will give, the financing a lender will provide, the insurance available, and the operator experience the project can recruit. A route that works technically but is not yet bankable at scale is a real option and a different kind of project.

The comparison should therefore ask, for each route, at what scale it has been demonstrated, what the reference plants are, what performance guarantees are offered, and what the project would need to do if the actual performance fell short of the design figure. Where the answer involves the project carrying technology risk it is not equipped to carry, that should be stated plainly.

Whichever route is selected, the output must be accepted by someone who will buy it for an application. That acceptance — qualification of the output for a specific use — is the gate that converts technical capability into revenue.

This is where routes differ in a way that technical comparison does not capture. An output that is chemically equivalent to virgin material at the required specification may be acceptable to a buyer without extensive qualification; one that is different in some respect may require the buyer to test, approve and possibly reformulate. That requirement is a cost and a lead time, and it can determine whether the project can sell what it produces within the period its business case assumes.

The practical discipline is to approach the qualification in parallel with the technology selection rather than after it. Identifying a potential buyer and establishing what they would need in order to accept the output — specifications, tests, trial volumes, contractual terms — is commercial work that informs the technical choice, and it is frequently the constraint that decides which route is realistic.

A completed comparison rarely identifies a single best route. It produces, for the specific input and required output, a shortlist of routes that can technically do the job, with each one's cost, maturity, residue and qualification profile stated, and the assumptions the conclusion rests on identified.

That is the useful output, because the decision among the shortlisted routes then turns on factors the technical comparison cannot resolve: the project's access to capital, its tolerance for technology risk, the buyers it can reach, the regulatory position in its market, and the scale of the plant it intends to build.

The comparison has done its job when it makes those trade-offs visible. It has failed when it produces a single answer — and the failure mode to avoid is applying a general conclusion about which technology is better to a specific input and output that the conclusion was not about.